518 



Mr. A. W. Ashton on the 

 Table IX. 





Duration (in 





No. of Coulombs. 





seconds) of 



Equation for Current. 





Date. 













Charge. 



Dis- 

 charge. 





Absorbed. 



Discharged. 



March 28. 



2400 





c = 61-2xl0-^-' 469 



7-19 Xl0~ 5 









66180 



c =87-5xl0-^-" 601 





18-4X10" 5 



March 29. 



1200 





c=62-7xl0-^-' 49i 



455X10- 









11520 



e=90-6xl0- 8 *-- 641 





7-25x10-° 





7200 





e=54-3xl0- 8 *-" 451 



1296 Xl0~ 5 









66360 



C -67-3xl0- 8 ^- 522 





28-4xl0~ 5 



March 30. 



1200 





c=58-2xl0- 8 ^- 458 



501 Xl0~ 5 









11940 



c=98-8xl0- 8 ^-" 6 ' 55 





7-30x10-° 







Total 



29-71 X 10- 5 



6135X10 -5 



13. In Table IX. an analysis of the quantities absorbed and 

 discharged by the cable during two days is given. The 

 quantity discharged has been calculated by integrating, for 

 the whole period during which the cable was short-circuited, 

 the equations representing the current. The quantity ab- 

 sorbed has been calculated on the assumption that the 

 variable part of the charging current represents the energy 

 which is being, as it were, " stored up" in the dielectric. It 

 will be seen that the quantity discharged is in each case 

 greater than the quantity absorbed during the corresponding 

 charging. Therefore, if the assumptions made are correct, 

 the dielectric, during these experiments, must on the whole 

 have discharged twdce the quantity it absorbed. On the 

 other hand, it may be that all the energy represented by the 

 constant current c is not being converted into heat during 

 the early period o£ electrification ; that is, the true con- 

 duction-current increases to c as electrification proceeds. 

 Again, the apparent discrepancy may be explained by the 

 fact that the previous treatment of the dielectric has resulted 

 in the storage of a comparatively large amount of energy 

 which is being given up while the cable is short-circuited, 

 and is augmenting the quantities discharged. In support of 

 this hypothesis, it has been found that dielectrics which 

 as far as could be ascertained had never been subjected to 

 electric stress, have yet given, when joined to an electro- 



